How to Design Adaptable Agents to Obtain a Consensus with Omoiyari

0Citations
Citations of this article
2Readers
Mendeley users who have this article in their library.

This article is free to access.

Abstract

This paper focuses on Omoiyari in Japanese as consideration/thoughtfulness for others in order to promote people to obtain a consensus among them especially in Internet society where is difficult to reach a consensus due to the limited communication/interaction, and aims at exploring the preliminary agent design that can promote people to obtain a consensus by Omoiyari. For this purpose, this paper starts by designing Omoiyari as the behaviors of filling the numerical and psychological gaps (e.g., a different income as the numerical gap while a different way of thinking among people as the psychological gap), and conducts the human subjective experiment to understand what kinds of aspects should be implemented in the Omaiyari agent. In detail, we employ Barnga as a cross-cultural game which cannot determine the winner without a consensus, and analyze the behaviors of the human players in Barnga with the emotional panels expressing happy, angry, sad, and surprise, which help the players to indirectly express their feeling to the other players. The analysis of human subject experiment has derived that the emotional panels are used to express their feeling for filling the numerical and psychological gaps and derive the change of the opponent’s behaviors. In detail, we found the following implications: (1) omoiyari-based behaviors are achieved by a sequence of showing the surprise/sad panels; showing the angry panel after recognizing the feeling of others; and changing the decision of the winner to the same one selected by others; (2) the surprise panel is increasingly used as the psychological gap increases; the sad panel is increasingly used as the numerical gap increases; the angry panel is used after recognizing the surprise/sad panels and contributes to changing the opponent’s behaviors; and the happy panel is used when the numerical and psychological gaps are filled.

Cite

CITATION STYLE

APA

Maekawa, Y., Uwano, F., Kitajima, E., & Takadama, K. (2019). How to Design Adaptable Agents to Obtain a Consensus with Omoiyari. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 11569 LNCS, pp. 462–475). Springer Verlag. https://doi.org/10.1007/978-3-030-22660-2_34

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free